Update changelog.

[youtubedl] / youtube_dl / extractor / twitch.py

diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py
index bbba394b0ede953f60c179970c8b41d4013f69c2..b39972b1efbfd46aab00a2f05e1a82605b8772b8 100644 (file)
--- a/youtube_dl/extractor/twitch.py
+++ b/youtube_dl/extractor/twitch.py
@@ -4,36 +4,41 @@ from __future__ import unicode_literals
 import itertools
 import re
 import random
 import itertools
 import re
 import random

+import json

 
 from .common import InfoExtractor
 from ..compat import (
 
 from .common import InfoExtractor
 from ..compat import (

-    compat_HTTPError,
+    compat_kwargs,

     compat_parse_qs,
     compat_str,
     compat_urllib_parse_urlencode,
     compat_urllib_parse_urlparse,
     compat_parse_qs,
     compat_str,
     compat_urllib_parse_urlencode,
     compat_urllib_parse_urlparse,

-    compat_urlparse,

 )
 from ..utils import (
     clean_html,
     ExtractorError,
 )
 from ..utils import (
     clean_html,
     ExtractorError,

+    float_or_none,

     int_or_none,
     int_or_none,

-    js_to_json,

     orderedSet,
     parse_duration,
     parse_iso8601,
     orderedSet,
     parse_duration,
     parse_iso8601,

+    qualities,
+    try_get,
+    unified_timestamp,

     update_url_query,
     update_url_query,

-    urlencode_postdata,
+    url_or_none,
+    urljoin,

 )
 
 
 class TwitchBaseIE(InfoExtractor):
 )
 
 
 class TwitchBaseIE(InfoExtractor):

-    _VALID_URL_BASE = r'https?://(?:www\.)?twitch\.tv'
+    _VALID_URL_BASE = r'https?://(?:(?:www|go|m)\.)?twitch\.tv'

 
     _API_BASE = 'https://api.twitch.tv'
     _USHER_BASE = 'https://usher.ttvnw.net'
 
     _API_BASE = 'https://api.twitch.tv'
     _USHER_BASE = 'https://usher.ttvnw.net'

-    _LOGIN_URL = 'http://www.twitch.tv/login'
-    _CLIENT_ID = 'jzkbprff40iqj646a697cyrvl0zt2m6'
+    _LOGIN_FORM_URL = 'https://www.twitch.tv/login'
+    _LOGIN_POST_URL = 'https://passport.twitch.tv/login'
+    _CLIENT_ID = 'kimne78kx3ncx6brgo4mv6wki5h1ko'

     _NETRC_MACHINE = 'twitch'
 
     def _handle_error(self, response):
     _NETRC_MACHINE = 'twitch'
 
     def _handle_error(self, response):


@@ -45,10 +50,11 @@ class TwitchBaseIE(InfoExtractor):
                 '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')),
                 expected=True)
 
                 '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')),
                 expected=True)
 

-    def _call_api(self, path, item_id, note):
+    def _call_api(self, path, item_id, *args, **kwargs):
+        kwargs.setdefault('headers', {})['Client-ID'] = self._CLIENT_ID

         response = self._download_json(
         response = self._download_json(

-            '%s/%s' % (self._API_BASE, path), item_id, note,
-            headers={'Client-ID': self._CLIENT_ID})
+            '%s/%s' % (self._API_BASE, path), item_id,
+            *args, **compat_kwargs(kwargs))

         self._handle_error(response)
         return response
 
         self._handle_error(response)
         return response
 


@@ -56,7 +62,7 @@ class TwitchBaseIE(InfoExtractor):
         self._login()
 
     def _login(self):
         self._login()
 
     def _login(self):

-        (username, password) = self._get_login_info()
+        username, password = self._get_login_info()

         if username is None:
             return
 
         if username is None:
             return
 


@@ -64,47 +70,64 @@ class TwitchBaseIE(InfoExtractor):
             raise ExtractorError(
                 'Unable to login. Twitch said: %s' % message, expected=True)
 
             raise ExtractorError(
                 'Unable to login. Twitch said: %s' % message, expected=True)
 

-        login_page, handle = self._download_webpage_handle(
-            self._LOGIN_URL, None, 'Downloading login page')
+        def login_step(page, urlh, note, data):
+            form = self._hidden_inputs(page)
+            form.update(data)

 
 

-        # Some TOR nodes and public proxies are blocked completely
-        if 'blacklist_message' in login_page:
-            fail(clean_html(login_page))
+            page_url = urlh.geturl()
+            post_url = self._search_regex(
+                r'<form[^>]+action=(["\'])(?P<url>.+?)\1', page,
+                'post url', default=self._LOGIN_POST_URL, group='url')
+            post_url = urljoin(page_url, post_url)

 
 

-        login_form = self._hidden_inputs(login_page)
+            headers = {
+                'Referer': page_url,
+                'Origin': page_url,
+                'Content-Type': 'text/plain;charset=UTF-8',
+            }

 
 

-        login_form.update({
-            'username': username,
-            'password': password,
-        })
+            response = self._download_json(
+                post_url, None, note, data=json.dumps(form).encode(),
+                headers=headers, expected_status=400)
+            error = response.get('error_description') or response.get('error_code')
+            if error:
+                fail(error)
+
+            if 'Authenticated successfully' in response.get('message', ''):
+                return None, None
+
+            redirect_url = urljoin(
+                post_url,
+                response.get('redirect') or response['redirect_path'])
+            return self._download_webpage_handle(
+                redirect_url, None, 'Downloading login redirect page',
+                headers=headers)

 
 

-        redirect_url = handle.geturl()
+        login_page, handle = self._download_webpage_handle(
+            self._LOGIN_FORM_URL, None, 'Downloading login page')

 
 

-        post_url = self._search_regex(
-            r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
-            'post url', default=redirect_url, group='url')
+        # Some TOR nodes and public proxies are blocked completely
+        if 'blacklist_message' in login_page:
+            fail(clean_html(login_page))

 
 

-        if not post_url.startswith('http'):
-            post_url = compat_urlparse.urljoin(redirect_url, post_url)
+        redirect_page, handle = login_step(
+            login_page, handle, 'Logging in', {
+                'username': username,
+                'password': password,
+                'client_id': self._CLIENT_ID,
+            })

 
 

-        headers = {'Referer': redirect_url}
+        # Successful login
+        if not redirect_page:
+            return

 
 

-        try:
-            response = self._download_json(
-                post_url, None, 'Logging in as %s' % username,
-                data=urlencode_postdata(login_form),
-                headers=headers)
-        except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
-                response = self._parse_json(
-                    e.cause.read().decode('utf-8'), None)
-                fail(response['message'])
-            raise
-
-        if response.get('redirect'):
-            self._download_webpage(
-                response['redirect'], None, 'Downloading login redirect page',
-                headers=headers)
+        if re.search(r'(?i)<form[^>]+id="two-factor-submit"', redirect_page) is not None:
+            # TODO: Add mechanism to request an SMS or phone call
+            tfa_token = self._get_tfa_info('two-factor authentication token')
+            login_step(redirect_page, handle, 'Submitting TFA token', {
+                'authy_token': tfa_token,
+                'remember_2fa': 'true',
+            })

 
     def _prefer_source(self, formats):
         try:
 
     def _prefer_source(self, formats):
         try:


@@ -151,6 +174,13 @@ class TwitchItemBaseIE(TwitchBaseIE):
         return self.playlist_result(entries, info['id'], info['title'])
 
     def _extract_info(self, info):
         return self.playlist_result(entries, info['id'], info['title'])
 
     def _extract_info(self, info):

+        status = info.get('status')
+        if status == 'recording':
+            is_live = True
+        elif status == 'recorded':
+            is_live = False
+        else:
+            is_live = None

         return {
             'id': info['_id'],
             'title': info.get('title') or 'Untitled Broadcast',
         return {
             'id': info['_id'],
             'title': info.get('title') or 'Untitled Broadcast',


@@ -161,6 +191,7 @@ class TwitchItemBaseIE(TwitchBaseIE):
             'uploader_id': info.get('channel', {}).get('name'),
             'timestamp': parse_iso8601(info.get('recorded_at')),
             'view_count': int_or_none(info.get('views')),
             'uploader_id': info.get('channel', {}).get('name'),
             'timestamp': parse_iso8601(info.get('recorded_at')),
             'view_count': int_or_none(info.get('views')),

+            'is_live': is_live,

         }
 
     def _real_extract(self, url):
         }
 
     def _real_extract(self, url):


@@ -209,7 +240,7 @@ class TwitchVodIE(TwitchItemBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
     _VALID_URL = r'''(?x)
                     https?://
                         (?:

-                            (?:www\.)?twitch\.tv/(?:[^/]+/v|videos)/|
+                            (?:(?:www|go|m)\.)?twitch\.tv/(?:[^/]+/v(?:ideo)?|videos)/|

                             player\.twitch\.tv/\?.*?\bvideo=v
                         )
                         (?P<id>\d+)
                             player\.twitch\.tv/\?.*?\bvideo=v
                         )
                         (?P<id>\d+)


@@ -262,6 +293,12 @@ class TwitchVodIE(TwitchItemBaseIE):
     }, {
         'url': 'https://www.twitch.tv/videos/6528877',
         'only_matching': True,
     }, {
         'url': 'https://www.twitch.tv/videos/6528877',
         'only_matching': True,

+    }, {
+        'url': 'https://m.twitch.tv/beagsandjam/v/247478721',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.twitch.tv/northernlion/video/291940395',
+        'only_matching': True,

     }]
 
     def _real_extract(self, url):
     }]
 
     def _real_extract(self, url):


@@ -350,9 +387,16 @@ class TwitchPlaylistBaseIE(TwitchBaseIE):
                 break
             offset += limit
         return self.playlist_result(
                 break
             offset += limit
         return self.playlist_result(

-            [self.url_result(entry) for entry in orderedSet(entries)],
+            [self._make_url_result(entry) for entry in orderedSet(entries)],

             channel_id, channel_name)
 
             channel_id, channel_name)
 

+    def _make_url_result(self, url):
+        try:
+            video_id = 'v%s' % TwitchVodIE._match_id(url)
+            return self.url_result(url, TwitchVodIE.ie_key(), video_id=video_id)
+        except AssertionError:
+            return self.url_result(url)
+

     def _extract_playlist_page(self, response):
         videos = response.get('videos')
         return [video['url'] for video in videos] if videos else []
     def _extract_playlist_page(self, response):
         videos = response.get('videos')
         return [video['url'] for video in videos] if videos else []


@@ -366,14 +410,17 @@ class TwitchProfileIE(TwitchPlaylistBaseIE):
     _VALID_URL = r'%s/(?P<id>[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
     _PLAYLIST_TYPE = 'profile'
 
     _VALID_URL = r'%s/(?P<id>[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
     _PLAYLIST_TYPE = 'profile'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'http://www.twitch.tv/vanillatv/profile',
         'info_dict': {
             'id': 'vanillatv',
             'title': 'VanillaTV',
         },
         'playlist_mincount': 412,
         'url': 'http://www.twitch.tv/vanillatv/profile',
         'info_dict': {
             'id': 'vanillatv',
             'title': 'VanillaTV',
         },
         'playlist_mincount': 412,

-    }
+    }, {
+        'url': 'http://m.twitch.tv/vanillatv/profile',
+        'only_matching': True,
+    }]

 
 
 class TwitchVideosBaseIE(TwitchPlaylistBaseIE):
 
 
 class TwitchVideosBaseIE(TwitchPlaylistBaseIE):


@@ -387,14 +434,17 @@ class TwitchAllVideosIE(TwitchVideosBaseIE):
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive,upload,highlight'
     _PLAYLIST_TYPE = 'all videos'
 
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive,upload,highlight'
     _PLAYLIST_TYPE = 'all videos'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'https://www.twitch.tv/spamfish/videos/all',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 869,
         'url': 'https://www.twitch.tv/spamfish/videos/all',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 869,

-    }
+    }, {
+        'url': 'https://m.twitch.tv/spamfish/videos/all',
+        'only_matching': True,
+    }]

 
 
 class TwitchUploadsIE(TwitchVideosBaseIE):
 
 
 class TwitchUploadsIE(TwitchVideosBaseIE):


@@ -403,14 +453,17 @@ class TwitchUploadsIE(TwitchVideosBaseIE):
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'upload'
     _PLAYLIST_TYPE = 'uploads'
 
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'upload'
     _PLAYLIST_TYPE = 'uploads'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'https://www.twitch.tv/spamfish/videos/uploads',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 0,
         'url': 'https://www.twitch.tv/spamfish/videos/uploads',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 0,

-    }
+    }, {
+        'url': 'https://m.twitch.tv/spamfish/videos/uploads',
+        'only_matching': True,
+    }]

 
 
 class TwitchPastBroadcastsIE(TwitchVideosBaseIE):
 
 
 class TwitchPastBroadcastsIE(TwitchVideosBaseIE):


@@ -419,14 +472,17 @@ class TwitchPastBroadcastsIE(TwitchVideosBaseIE):
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive'
     _PLAYLIST_TYPE = 'past broadcasts'
 
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive'
     _PLAYLIST_TYPE = 'past broadcasts'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'https://www.twitch.tv/spamfish/videos/past-broadcasts',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 0,
         'url': 'https://www.twitch.tv/spamfish/videos/past-broadcasts',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 0,

-    }
+    }, {
+        'url': 'https://m.twitch.tv/spamfish/videos/past-broadcasts',
+        'only_matching': True,
+    }]

 
 
 class TwitchHighlightsIE(TwitchVideosBaseIE):
 
 
 class TwitchHighlightsIE(TwitchVideosBaseIE):


@@ -435,14 +491,17 @@ class TwitchHighlightsIE(TwitchVideosBaseIE):
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'highlight'
     _PLAYLIST_TYPE = 'highlights'
 
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'highlight'
     _PLAYLIST_TYPE = 'highlights'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'https://www.twitch.tv/spamfish/videos/highlights',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 805,
         'url': 'https://www.twitch.tv/spamfish/videos/highlights',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 805,

-    }
+    }, {
+        'url': 'https://m.twitch.tv/spamfish/videos/highlights',
+        'only_matching': True,
+    }]

 
 
 class TwitchStreamIE(TwitchBaseIE):
 
 
 class TwitchStreamIE(TwitchBaseIE):


@@ -450,7 +509,7 @@ class TwitchStreamIE(TwitchBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
     _VALID_URL = r'''(?x)
                     https?://
                         (?:

-                            (?:www\.)?twitch\.tv/|
+                            (?:(?:www|go|m)\.)?twitch\.tv/|

                             player\.twitch\.tv/\?.*?\bchannel=
                         )
                         (?P<id>[^/#?]+)
                             player\.twitch\.tv/\?.*?\bchannel=
                         )
                         (?P<id>[^/#?]+)


@@ -481,6 +540,12 @@ class TwitchStreamIE(TwitchBaseIE):
     }, {
         'url': 'https://player.twitch.tv/?channel=lotsofs',
         'only_matching': True,
     }, {
         'url': 'https://player.twitch.tv/?channel=lotsofs',
         'only_matching': True,

+    }, {
+        'url': 'https://go.twitch.tv/food',
+        'only_matching': True,
+    }, {
+        'url': 'https://m.twitch.tv/food',
+        'only_matching': True,

     }]
 
     @classmethod
     }]
 
     @classmethod


@@ -566,21 +631,23 @@ class TwitchStreamIE(TwitchBaseIE):
         }
 
 
         }
 
 

-class TwitchClipsIE(InfoExtractor):
+class TwitchClipsIE(TwitchBaseIE):

     IE_NAME = 'twitch:clips'
     _VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
     _TESTS = [{
     IE_NAME = 'twitch:clips'
     _VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
     _TESTS = [{

-        'url': 'https://clips.twitch.tv/ea/AggressiveCobraPoooound',
+        'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat',

         'md5': '761769e1eafce0ffebfb4089cb3847cd',
         'info_dict': {
         'md5': '761769e1eafce0ffebfb4089cb3847cd',
         'info_dict': {

-            'id': 'AggressiveCobraPoooound',
+            'id': '42850523',

             'ext': 'mp4',
             'title': 'EA Play 2016 Live from the Novo Theatre',
             'thumbnail': r're:^https?://.*\.jpg',
             'ext': 'mp4',
             'title': 'EA Play 2016 Live from the Novo Theatre',
             'thumbnail': r're:^https?://.*\.jpg',

+            'timestamp': 1465767393,
+            'upload_date': '20160612',

             'creator': 'EA',
             'uploader': 'stereotype_',
             'creator': 'EA',
             'uploader': 'stereotype_',

-            'uploader_id': 'stereotype_',
+            'uploader_id': '43566419',

         },
     }, {
         # multiple formats
         },
     }, {
         # multiple formats


@@ -591,34 +658,63 @@ class TwitchClipsIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 

-        webpage = self._download_webpage(url, video_id)
-
-        clip = self._parse_json(
-            self._search_regex(
-                r'(?s)clipInfo\s*=\s*({.+?});', webpage, 'clip info'),
-            video_id, transform_source=js_to_json)
+        status = self._download_json(
+            'https://clips.twitch.tv/api/v2/clips/%s/status' % video_id,
+            video_id)

 
 

-        title = clip.get('channel_title') or self._og_search_title(webpage)
+        formats = []

 
 

-        formats = [{
-            'url': option['source'],
-            'format_id': option.get('quality'),
-            'height': int_or_none(option.get('quality')),
-        } for option in clip.get('quality_options', []) if option.get('source')]
-
-        if not formats:
-            formats = [{
-                'url': clip['clip_video_url'],
-            }]
+        for option in status['quality_options']:
+            if not isinstance(option, dict):
+                continue
+            source = url_or_none(option.get('source'))
+            if not source:
+                continue
+            formats.append({
+                'url': source,
+                'format_id': option.get('quality'),
+                'height': int_or_none(option.get('quality')),
+                'fps': int_or_none(option.get('frame_rate')),
+            })

 
         self._sort_formats(formats)
 
 
         self._sort_formats(formats)
 

-        return {
-            'id': video_id,
-            'title': title,
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'creator': clip.get('broadcaster_display_name') or clip.get('broadcaster_login'),
-            'uploader': clip.get('curator_login'),
-            'uploader_id': clip.get('curator_display_name'),
+        info = {

             'formats': formats,
         }
             'formats': formats,
         }

+
+        clip = self._call_api(
+            'kraken/clips/%s' % video_id, video_id, fatal=False, headers={
+                'Accept': 'application/vnd.twitchtv.v5+json',
+            })
+
+        if clip:
+            quality_key = qualities(('tiny', 'small', 'medium'))
+            thumbnails = []
+            thumbnails_dict = clip.get('thumbnails')
+            if isinstance(thumbnails_dict, dict):
+                for thumbnail_id, thumbnail_url in thumbnails_dict.items():
+                    thumbnails.append({
+                        'id': thumbnail_id,
+                        'url': thumbnail_url,
+                        'preference': quality_key(thumbnail_id),
+                    })
+
+            info.update({
+                'id': clip.get('tracking_id') or video_id,
+                'title': clip.get('title') or video_id,
+                'duration': float_or_none(clip.get('duration')),
+                'views': int_or_none(clip.get('views')),
+                'timestamp': unified_timestamp(clip.get('created_at')),
+                'thumbnails': thumbnails,
+                'creator': try_get(clip, lambda x: x['broadcaster']['display_name'], compat_str),
+                'uploader': try_get(clip, lambda x: x['curator']['display_name'], compat_str),
+                'uploader_id': try_get(clip, lambda x: x['curator']['id'], compat_str),
+            })
+        else:
+            info.update({
+                'title': video_id,
+                'id': video_id,
+            })
+
+        return info